算力网络部署方案分析

doi:10.11959/j.issn.1000-0801.2022134

摘要/Abstract

摘要：

国家对算力网络的部署提出了更高要求，电信运营商加速落实算网融合工作。为解决计算和网络相互协同、调度问题，在对算力网络相关标准研究的基础上，总结了算力网络研究进展，分析了算力网络业务需求，结合电信运营商现网实际，从算力储备、传输网络、网络路由、协同调度 4 个方面，提出了算力网络可持续性建设部署方案，为算力网络建设初期提供参考。

关键词: 算力网络, 路由, 调度, 优化

Abstract:

The government puts forward higher requirements for the deployment of computing force network (CFN), so the telecom operators accelerate the implementation of computing network integration.Aiming to solve the problems of computing and network coordination and scheduling, based on the study of standards related to CFN, the research progress of CFN was summarized, the business requirements of CFN were analyzed, combined with the current reality of telecom operators, the deployment scheme of sustainable construction of CFN was put forward, which concerned with computing power reserve, transmission network, network routing and intelligent scheduling, it provides a reference for the early stage of CFN construction.

Key words: computing force network, routing, scheduling, optimization

中图分类号:

TN914

王丽莉. 算力网络部署方案分析[J]. 电信科学, 2022, 38(6): 172-180.

Lili WANG. Construction scheme analysis of computing force network[J]. Telecommunications Science, 2022, 38(6): 172-180.

图/表 9

图1

图2

图3

表1

表2

表3

图4

表4

图5

参考文献 18

[1]	国家发展和改革委员会, 中共中央网络安全和信息化委员办公室, 工业和信息化部,等. 全国一体化大数据中心协同创新体系算力枢纽实施方案[R]. 2021.
	National Development and Reform Commission, office of central cyberspace Affairs Commission, Ministry of Industry and Information Technology of the Reople's Republic of China, et al. Implementation plan for computing hub of collaborative innovation system of national integrated big data center[R]. 2021.
[2]	中国移动. 算力网络白皮书[R]. 2021.
	China Mobile Communications Group Co.,Ltd.. White paper of computing power network[R]. 2021.
[3]	中国联通. 中国联通算力网络白皮书[R]. 2019.
	China Unicom. White paper of China Unicom computing power network[R]. 2019.
[4]	姚惠娟, 陆璐, 段晓东 . 算力感知网络架构与关键技术[J]. 中兴通讯技术, 2021,27(3): 7-11.
	YAO H J , LU L , DUAN X D . Architecture and key technologies for computing-aware networking[J]. ZTE Technology Journal, 2021,27(3): 7-11.
[5]	蒋林涛 . 云计算、边缘计算和算力网络[J]. 信息通信技术, 2020,14(4): 4-8.
	JIANG L T . Cloud computing,edge computing and computing power network[J]. Information and Communications Technologies, 2020,14(4): 4-8.
[6]	何涛, 曹畅, 唐雄燕 ,等. 面向 6G 需求的算力网络技术[J]. 移动通信, 2020,44(6): 131-135.
	HE T , CAO C , TANG X Y ,et al. Research on computing power network technology for 6G requirements[J]. Mobile Communications, 2020,44(6): 131-135.
[7]	雷波, 陈运清 . 边缘计算与算力网络:5G+AI 时代的新型算力平台与网络连接[J]. 中国信息化, 2020(12): 113.
	LEI B , CHEN Y Q . Edge computing and computing power network:a new computing power platform and network connection in 5G+AIera[J]. Zhongguo Xinxihua, 2020(12): 113.
[8]	黄光平, 罗鉴, 周建锋 . 算力网络架构与场景分析[J]. 信息通信技术, 2020,14(4): 16-22.
	HUANG G P , LUO J , ZHOU J F . Analysis of computation network architecture and according scenarios[J]. Information and Communications Technologies, 2020,14(4): 16-22.
[9]	段晓东, 姚惠娟, 付月霞 ,等. 面向算网一体化演进的算力网络技术[J]. 电信科学, 2021,37(10): 76-85.
	DUAN X D , YAO H J , FU Y X ,et al. Computing force network technologies for computing and network integration evolution[J]. Telecommunications Science, 2021,37(10): 76-85.
[10]	李铭轩, 曹畅, 唐雄燕 ,等. 面向算力网络的边缘资源调度解决方案研究[J]. 数据与计算发展前沿, 2020,2(4): 80-91.
	LI M X , CAO C , TANG X Y ,et al. Research on edge resource scheduling solutions for computing power network[J]. Frontiers of Data ＆ Computing, 2020,2(4): 80-91.
[11]	狄筝, 曹一凡, 仇超 ,等. 新型算力网络架构及其应用案例分析[J]. 计算机应用, 2022: 1-7.
	DI Z , CAO Y F , QIU C ,et al. New computing power network architecture and its application case analysis[J]. Journal of Computer Application, 2022: 1-7.
[12]	张帅, 曹畅, 唐雄燕 . 基于SRv6的算力网络技术体系研究[J]. 中兴通讯技术, 2022,28(1): 11-15.
	ZHANG S , CAO C , TANG X Y . Computing power network technology architecture based on SRv6[J]. ZTE Technology Journal, 2022,28(1): 11-15.
[13]	黄光平, 史伟强, 谭斌 . 基于SRv6的算力网络资源和服务编排调度[J]. 中兴通讯技术, 2021,27(3): 23-28.
	HUANG G P , SHI W Q , TAN B . Computing power network resources based on SRv6 and its service arrangement and sche-duling[J]. ZTE Technology Journal, 2021,27(3): 23-28.
[14]	李铭轩, 常培, 崔童 ,等. 面向 FaaS 的算网异构资源调度技术[J]. 信息通信技术, 2021,15(4): 44-49,58.
	LI M X , CHANG P , CUI T ,et al. FaaS-oriented computing network heterogeneous resource scheduling technology[J]. In-formation and Communications Technologies, 2021,15(4): 44-49,58.
[15]	孙钰坤, 张兴, 雷波 . 边缘算力网络中智能算力感知路由分配策略研究[J]. 无线电通信技术, 2022,48(1): 60-67.
	SUN Y K , ZHANG X , LEI B . Study on intelligent computing aware route allocation policy in edge computing-aware net-works[J]. Radio Communications Technology, 2022,48(1): 60-67.
[16]	邱勤, 徐天妮, 于乐 ,等. 算力网络安全架构与数据安全治理技术[J]. 信息安全研究, 2022,8(4): 340-350.
	QIU Q , XU T N , YU L ,et al. Computing force network security architecture and data security governance technology[J]. Journalof Information Security Research, 2022,8(4): 340-350.
[17]	温瑶, 陆晶晶, 卢华 ,等. 融合区块链的算力网络信任评估与保障方案研究[J]. 南京邮电大学学报（自然科学版）, 2021,41(4): 99-106.
	WEN Y , LU J J , LU H ,et al. Blockchain-based trust evaluation and guarantee scheme for computing power network[J]. Journal of Nanjing University of Posts and Telecommunications (Natu-ral Science Edition), 2021,41(4): 99-106.
[18]	ITU-T. Computing power network-framework and architecture:Y.2501[S]. 2021.

节点层级	主要业务特征	传输技术方案选择建议
国家算力枢纽节点	超大带宽、超低时延	400 GB/200 GB OTN、OXC全光交叉连接
省级算力节点
地市/县算力节点	大带宽、低时延	400 GB/200 GB OTN、OXC全光交叉连接
边缘算力节点	较大带宽、较低时延	200 GB/100 GB SPN、FlexE切片

段落			内部动因				外部路由因素		分值小计
段落	纤芯亟待扩容	第二主用路由	属于多环共臂段落	可储备过江资源储备	一二干可协同解决	自有硅管路由	租用需协调路由	暂不具备路由条件	分值小计
分值	3	3	2	2	2	3	-3	-5
A市-B市	√		√		√	√			10
C市-D市	√					√			6
A市-E市		√	√					√	0
E市-F市		√	√	√			√	√	-1
F市-G市		√			√		√	√	-3

目的资源池	时延/ms	计算资源总量/vCPU	存储资源总量/TB	空闲计算资源占比	空闲存储资源占比
池C1-1	1	20	20	50%	50%
池C1-2	2	20	20	50%	50%
池C1	2	50	50	50%	50%
池D1	4	50	50	50%	40%
池D1-1	5	20	20	50%	50%
池D1-2	5	20	20	50%	50%
...	...	...	...	...	...

业务源IP地址	目的资源池IP地址	下一跳接口/IP地址	时延/ms	需要使用的计算资源/vCPU	需要使用的存储资源/TB
—	池C1-1	—	1	5	5